SABnzbd Forums

I am using sabnzbd on ubuntu 12.10. It runs fine on most nzb files but some are not finishing. They are also not going to the history but just remain in the queue with 6 or 5 mb remaining.
How can I solve this? Or what information do you need to get an idea of my problem? (i am very new to all this).

Thx in advance.

Which SABnzbd? Try to upgrade to the latest version (0.7.6).
See our Ubuntu/Debian board on how to get recent packages.

Hello,
I am already running 0.7.6

You didn't tell.
Do you have "limited retries" enabled in Config->Switches?
Do you have "only for optional servers" off?
Do you have any "optional" servers in Config->Servers.
Some servers simply refuse to admit they're missing an article.
Or it takes them longer than the timeout value that you set for that server.
Using "limited retries", will break the deadlock of eternal retries.

Limited retries is on 3
"only for optional servers" is off
I do not have any optional servers

I put a copy of my logfile on pastebin. Perhaps it will help.
http://pastebin.com/QE0184Zc

What kind of files still need to be downloaded?
(Click on the queue entry).

They are par2 files.

Strange.
What happens if you change the priority of the stuck job to "Stop"?

It goed into post processing for a few seconds and then jumps back into the queue

edit: After rebooting my system one of them sais it is failed with the message: » Invalid par2 files, cannot verify or repair

Please email the NZB to bugs@sabnzbd.org
Add the URL of this post and which Usenet provider you're using.

Hi,

I have exactly the same problem.
Download hangs for the last few MB.
Also the PAR2 files.

The usenetserver I have is I-Telligent.

Is your issue solved?

thoraxx wrote: I have exactly the same problem.

That's what they all say
Are you also using release 0.7.6 and do you have a specific NZB the shows this behaviour?

Yes using SABnzbd Version: 0.7.6

No specific NZB which fails. But after some analysis I see that OLD NZB's close to my usenetserver retention are failing the most.

I am now trying Supernews as usenetserver.
So far so good, no fails yet.
Maybe this is not a SABnzbd related error, but just my usenetserver retention!

Ill keep you updated.

The scenario you're talking about is familiar.
It means that your server isn't responding (or not fast enough) to requests for older articles.
This may be worse for missing articles.
So there are a few important settings for this:
Your server timeout: don't make it too short. Choose 120 rather than 30.
Retries (Config->Switches) make sure it has a value, preferably the default 3.
Below "Retries" is "Only for optional servers", make sure it's disabled.
Even so three retries of 120 seconds means that it's still stuck for 6 minutes at least.
That's the price...

BTW: a too short timeout and unlimited retries will cause a deadlock for the job.